# vlt5-base-keywords by VoiceLab.ai: Open-Weight Model Source: https://savrn.com/models/vlt5-base-keywords Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve vlt5-base-keywords (275M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 0.6 GB | 0.7 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.3 GB | 0.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.1 GB | 0.2 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 23, 2026. [vlt5-base-keywords on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/vlt5-base-keywords/gpus) ## SAVRN's Notes on vlt5-base-keywords Three to five keywords from an abstract-length piece of text, extractive or abstractive, is what VoiceLab.ai built this 275M-parameter T5 to return. At 16-bit it needs 0.7 GB of memory, and 0.3 GB at 8-bit, so the cheapest listed setup, one 192 GB MI300X at $1.85 per hour on-demand, is overkill for it alone. Its natural home is leftover memory on a card already busy with other work. Longer documents must be split into chunks, and the training set is posmac. CC BY 4.0 asks two things in return for commercial use: credit the creator and indicate what you changed, which matters once you fine-tune it on your own tags. Settle the chunk size before committing, since output length and keyword count follow the training data and no context length is given. Weights are stored in float32, the last update was August 16, 2023, and the method is in arXiv:2209.14008. SAVRN Research, 2026-09-18 ## Model Card By VoiceLab.ai, published under cc-by-4.0, revision 89a069b6626a. ### Keyword Extraction from Short Texts with T5 Our vlT5 model is a keyword generation model based on encoder-decoder architecture using Transformer blocks presented by Google ([https://huggingface.co/t5-base](https://huggingface.co/t5-base)). The vlT5 was trained on scientific articles corpus to predict a given set of keyphrases based on the concatenation of the article’s abstract and title. It generates precise, yet not always complete keyphrases that describe the content of the article based only on the abstract. Keywords generated with vlT5-base-keywords: encoder-decoder architecture, keyword generation Results on demo model (different generation method, one model per language): Our vlT5 model is a keyword generation model based on encoder-decoder architecture using Transformer blocks presented by Google ([https://huggingface.co/t5-base](https://huggingface.co/t5-base)). The vlT5 was trained on scientific articles corpus to predict a given set of keyphrases based on the concatenation of the article’s abstract and title. It generates precise, yet not always complete keyphrases that describe the content of the article based only on the abstract. [Read the full model card (900 words)](https://savrn.com/models/vlt5-base-keywords/card) ## Configuration Architecture T5ForConditionalGeneration Vocabulary size 50,048 Stored precision float32 Model type t5 ## Identity and Version Repository Voicelab/vlt5-base-keywords Publisher VoiceLab.ai Task Text generation Modality Text Library transformers Parameters 275M parameters Languages pl, en Revision 89a069b6626ad817ceaddf1f08271cd1aadb138a First published 2022-09-27 Last updated 2023-08-16 ## Files and Weights 8 files, 2.2 GB in total. The weights are 2 files totalling 2.2 GB in bin, safetensors. Weights2 files · 2.2 GB Configuration2 files · 881 B Tokenizer2 files · 1.1 MB Documentation1 file · 11.2 KB Repository1 file · 1.4 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.safetensors | Weights | 1.1 GB | c8d9cefb35a2 | | pytorch_model.bin | Weights | 1.1 GB | 2b055a39ebd3 | | config.json | Configuration | 807 B | — | | special_tokens_map.json | Configuration | 74 B | — | | README.md | Documentation | 11.2 KB | — | | .gitattributes | Repository | 1.4 KB | — | | spiece.model | Tokenizer | 1.1 MB | a200b4b9d710 | | tokenizer_config.json | Tokenizer | 265 B | — | ## License and Download License cc-by-4.0 Access Open weights, no gate Download size 2.2 GB [Download from VoiceLab.ai](https://huggingface.co/Voicelab/vlt5-base-keywords) Released by VoiceLab.ai through its official repository on Hugging Face. [Read the license](https://creativecommons.org/licenses/by/4.0/). ## Built From - Described by arXiv:2209.14008 - Trained on (disclosed) posmac ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 2.2 GB | | 16-bit | 0.6 GB | | 8-bit | 0.3 GB | | 4-bit | 0.1 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Compare vlt5-base-keywords - [pythia-160m vs vlt5-base-keywords](https://savrn.com/models/compare/pythia-160m-vs-vlt5-base-keywords) ## Questions About vlt5-base-keywords ### How much GPU memory does vlt5-base-keywords need? About 0.7 GB at 16-bit and 0.2 GB at 4-bit: the weights (275M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run vlt5-base-keywords on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### Can I use vlt5-base-keywords commercially? Yes. vlt5-base-keywords is released under Creative Commons Attribution 4.0. CC BY 4.0 permits sharing and adapting the work, including commercially, provided the creator is credited and changes are indicated. ## Similar Models Model · Text generation ### [Gemma-3-270m-Opus-Distil](https://savrn.com/models/gemma-3-270m-opus-distil) [Convergent Intelligence](https://savrn.com/model-publishers/reaperdoesntknow) This model is a fine-tuned derivative of google/gemma-3-270m, adapted using the Convergent Intelligence sparse fine-tuning setup originally tested on Liquid Foundation Models. The checkpoint was trained on reasoning-style English examples from angrygiraffe/claude-opus-4.6-4.7-reasoning-8.7k using a targeted adaptation strategy and the custom CIxOpt optimizer framework. The goal of this model is to test whether a compact Gemma 3 270M backbone can be shaped toward reasoning-style text generation through selective parameter participation rather than broad full-model modification. This is an experimental research checkpoint intended for evaluation, local testing, optimizer research, and… Open weights gemma 268M parameters 262,144 tokens transformers [View model](https://savrn.com/models/gemma-3-270m-opus-distil) Model · Text generation ### [gemma-270m-math-reasoner](https://savrn.com/models/gemma-270m-math-reasoner) [Convergent Intelligence](https://savrn.com/model-publishers/reaperdoesntknow) This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019). Open weights apache-2.0 268M parameters 32,768 tokens transformers [View model](https://savrn.com/models/gemma-270m-math-reasoner) Model · Text generation ### [outward-gemma3-270m](https://savrn.com/models/outward-gemma3-270m) [Harsh tuli](https://savrn.com/model-publishers/harsh452) The on-device storyteller for Outward, an offline walking adventure. You go outside, find something real ("rainwater running down a drain"), and this model writes the next 60–120 words of the story around it. It runs entirely on the phone, in the browser, with llama.cpp (wllama). Modified by the Outward project. This is a Model Derivative of google/gemma-3-270m-it: fully fine-tuned on 3,186 stories written by a LoRA fine-tuned Qwen3.5-4B teacher (trained on Tinker), then converted to GGUF and quantized to Q40. The eval set has 50 held-out prompts × 3 samples, temperature 0.7, scored with the automatic rubric in the Outward repo: uses the discovery, 60–120 words, safe and outdoors (risky… Open weights gemma 268M parameters 32,768 tokens gguf [View model](https://savrn.com/models/outward-gemma3-270m) Model · Text generation ### [gemma-3-270m-african-diacritizer](https://savrn.com/models/gemma-3-270m-african-diacritizer) [Aletheia AI](https://savrn.com/model-publishers/aletheiaaing) Open weights gemma 268M parameters transformers [View model](https://savrn.com/models/gemma-3-270m-african-diacritizer) Model · Text generation ### [Greta-0.3b-Preview](https://savrn.com/models/greta-0-3b-preview) [Garbo AI AB](https://savrn.com/model-publishers/garboab) This model is currently still in post-training and is pre-released as a preview checkpoint with gated access only. Access requested at publisher 262M parameters transformers [View model](https://savrn.com/models/greta-0-3b-preview) Model · Text generation ### [PMA-1.3-Rosa-256M](https://savrn.com/models/pma-1-3-rosa-256m) [PatriotMemory-AI](https://savrn.com/model-publishers/patriotmemory-ai) Rosa is Patriot Memory's 253M-parameter edge assistant for English and Traditional Chinese — built to fit the edge devices you actually ship, with a vocabulary trained natively on Traditional Chinese. It provides accurate information regarding: If you run into multi-GPU tensor device mismatch errors: RuntimeError: Expected all tensors to be on the same device... Run the script with CUDAVISIBLEDEVICES=0 to isolate execution to GPU 0. Good: introducing herself in both languages; answering arithmetic word problems with visible English reasoning steps; clean Traditional Chinese — glyphs, vocabulary and register; running fully offline on edge hardware. Not: her reasoning traces are formatted… Open weights apache-2.0 262M parameters 4,096 tokens transformers [View model](https://savrn.com/models/pma-1-3-rosa-256m) ## VoiceLab.ai [All models and datasets](https://savrn.com/model-publishers/voicelab) ## Versions - [89a069b6626a](https://savrn.com/models/vlt5-base-keywords/versions/89a069b6626a) · current 2026-09-18 ## Explore More - [All text generation models](https://savrn.com/models/tasks/text-generation) - [All models under cc-by-4.0](https://savrn.com/models/licenses/cc-by-4-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-20. - [Hugging Face record](https://huggingface.co/Voicelab/vlt5-base-keywords) - [How the hub is built](https://savrn.com/model-hub/methodology)